/deadparrot-ml

An introduction to Machine Learning as part of the Lowell Observatory Dead Parrots Seminar

Primary LanguageJupyter Notebook

An Introduction to Machine Learning

This seminar series takes place in the framework of the Lowell Observatory Dead Parrots Python Seminar.

Slides and example notebooks will be provided here.

Schedule

This schedule is preliminary and might be subject to changes.

Nov 15 - a gentle introduction

  • what is machine learning, what is AI?
  • tasks: unsupervised learning/supervised learning
  • software: scikit-learn (and a little bit of pytorch)
  • some cool application examples as motivation

Nov 22 - unsupervised learning

  • k-means clustering
  • kernel density estimation
  • principal component analysis

Nov 25 - supervised learning basics and concepts

  • data: training data, test data, iid
  • objective functions
  • metrics and errors
  • generalization and regularization
  • parameters and hyperparameters

Dec 4: more supervised learning: methods

  • k-nearest neighbors: classification and regression
  • decision trees
  • ensemble methods: random forests
  • hyperparameter tuning

Dec 13: deep learning and neural networks

  • neurons and neural networks
  • perceptrons and multi-layer perceptrons
  • stochastic gradient descent and backpropagation
  • convolutional neural networks
  • deep learning examples